|Home | About | Journals | Submit | Contact Us | Français|
Analysis of tissues from cancers, precancers, and normal tissues provides a means to identify candidate markers for disease detection. The only proteomic technology platform capable of large-scale inventory and identification of serum proteins is shotgun proteomics, in which proteins are first digested to peptides and then the peptides are subjected to analysis by multidimensional liquid chromatography-tandem MS (LC-MS-MS). However, current implementations of shotgun proteome analyses are limited in both sample throughput and reproducibility in identification and detection, particularly for lower-abundance proteins. Here, we describe efforts to refine, standardize, and implement shotgun proteomics platforms for application to high-throughput analysis of clinical tissue specimens. Our guiding principles in developing and standardizing shotgun proteome analysis platforms, in order of decreasing priority, are to (1) achieve sufficient reproducibility to allow single analyses to replace multiple replicates, (2) reduce the amount of MS instrument time required for analysis, thus increasing throughput, and (3) achieve the greatest sensitivity and depth of coverage possible, with the ultimate goal of equaling or exceeding the performance of lower-throughput shotgun proteome analyses in current use. Refinement of the multidimensional LC-MS-MS platform is focused on (1) improving the reproducibility and standardization of peptide separations by replacing strong cation exchange separations with isoelectric focusing on immobilized pH gradient strips; (2) employing new methods to acquire MS-MS spectra in LC-MS-MS analyses using hybrid LTQ-Orbitrap instruments; (3) applying new data-analysis algorithms and software to identify peptides and proteins from MS-MS data and to quantify with label-free methods. A major challenge is the statistical comparison of multiple complex datasets derived by shotgun analyses to identify tissue-specific proteomic characteristics that can be selected as candidate markers.